Function to plot heatmaps separating groups generated by SOM or k-means

Description

This function display individual heatmaps for each group constructed by divisive clustering techniques, like SOM or k-means. It is possible to add hierarchical dendrograms to each group individually.

Usage

1
2
heatmapsM(data, groups, sampleT=NULL, doHier=FALSE, distfun=dist,
          hclustfun=hclust, ...)

Arguments

data

numeric matrix to be displayed, including all observations from all groups.

groups

numeric or character vector (with length equal to columns or rows of data) with identification of the groups to be separated. If numeric must contain the indexes and if character must contain the colnames or rownames.

sampleT

list with 2 vectors, the first one specifying the first character of each sample label to be coloured according to the colours specified in the second vector.

doHier

logical indicating to plot or not the hierarchical branch in the other dimension of the matrix.

distfun

the function to be used for distance calculation.

hclustfun

the function for to be used for hierarchical cluster calculation.

...

additional parameters for image function.

Details

This function is used internally by the functions hierM, somM and kmeansM. If you want to the hierarchical dendrogram in the other dimension, pay attention to the number of observations in this dimension, because the calculation of the hierarchical branch may be very time consuming!

Value

This function display the heatmaps and don't return any object or value.

Author(s)

Elier B. Cristo, addapted by Gustavo H. Esteves <gesteves@vision.ime.usp.br>

See Also

hierM, somM, kmeansM

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.